Hello! 👋

I'm Haley Li

Senior Research Engineer

About Me

I am a senior research engineer at Huawei Canada's Intelligent Cloud Infrastructure Lab. My current focus is on LLM inference optimization and asynchronous RL fine-tuning for large-scale MoE models. I previously completed my M.Sc in computer science at the University of British Columbia, where I applied reinforcement learning techniques to load balancing and scheduling for serverless and search engines.

Python Java JavaScript TypeScript PyTorch Git Machine Learning Reinforcement Learning vLLM AReaL Ray Docker Kubernetes

Publications

01

ElasticMoE: An Efficient Auto Scaling Method for Mixture-of-Experts Models

Gursimran Singh, Timothy Yu, Haley Li, Cheng Chen, Hanieh Sadri, Qintao Zhang, Yu Zhang, Ying Xiong, Yong Zhang, and Zhenan Fan. ElasticMoE: An Efficient Auto Scaling Method for Mixture-of-Experts Models. arXiv.

02

xDeepServe: Model-as-a-Service on Huawei CloudMatrix384

XDS Team @ Huawei. 2025. xDeepServe: Model-as-a-Service on Huawei CloudMatrix384. arXiv.

03

Augmenting Operations Research with Auto-Formulation of Optimization Models from Problem Descriptions

Rindranirina Ramamonjison, Haley Li, Timothy T. Yu, Shiqi He, Vishnu Rengan, Amin Banitalebi-Dehkordi, Zirui Zhou, and Yong Zhang. 2022. Augmenting Operations Research with Auto-Formulation of Optimization Models from Problem Descriptions. Proceedings of EMNLP 2022 Industry Track.

04

NL4Opt Competition: Formulating Optimization Problems Based on Their Natural Language Descriptions

Rindranirina Ramamonjison, Timothy T. Yu, Raymond Li, Haley Li, Giuseppe Carenini, Bissan Ghaddar, Shiqi He, Mahdi Mostajabdaveh, Amin Banitalebi-Dehkordi, Zirui Zhou, and Yong Zhang. 2023. NL4Opt Competition: Formulating Optimization Problems Based on Their Natural Language Descriptions. Proceedings of the NeurIPS 2022 Competitions Track

05

CoSpot: A Cooperative VM Allocation Framework for Increased Revenue from Spot Instances

Syed M. Iqbal, Haley Li, Shane Bergsma, Ivan Beschastnikh, Alan J. Hu. 2022. CoSpot: A Cooperative VM Allocation Framework for Increased Revenue from Spot Instances. SoCC '22: Proceedings of the 13th Symposium on Cloud Computing.

Patents

01

Methods and systems for model generation and instantiation of optimization models from markup documents

WO2024092553A1

Experience

Senior Research Engineer

Huawei Canada

Feb 2025 - Present
  • Intelligent Cloud Infrastructure Lab
  • Large language model inference optimization
  • Fault tolerant inference serving
  • Asynchronous reinforcement fine-tuning for large-scale MoE

Research Assistant

University of British Columbia

Sep 2021 - Dec 2023
  • Systopia Lab
  • Applications of RL in cloud scheduling and load baalancing

AI Researcher Intern

Huawei Canada

May 2022 - Sep 2022
  • Big Data and Intelligence Platform Lab
  • Fine-tuning language models for linear programming solver workflows

Undergraduate Researcher

University of British Columbia

May 2020 - Aug 2021
  • Systopia Lab
  • Applications of ML techniques in cloud scheduling

Developer

Hypatia Systems

Jan 2019 - Jan 2020
  • Frontend development for Electron-based LaTeX editing application

Education

Master of Science in Computer Science

University of British Columbia

2021 - 2024

    Bachelor of Science in Computer Science and Mathematics

    University of British Columbia

    2016 - 2021